modèles/AutismMix SDXL - AutismMix_DPO

AutismMix SDXL - AutismMix_DPO

11/9/2025

5:36:23 PM

Mots-clés et tags associés

anime,anime style,autismix_anon,autismmix,autismmix_confetti,autismmix_dpo,autismmix_pony,autismmix sdxl,modèle de base,checkpoint,checkpoint xl,lora compatibility,model,pony diffusion model,ponyxl,sdxl 1.0,sdxl anime,stable diffusion model

Nero Claudius sitting confidently on a throne in a regal castle room, legs crossed, smirking with a cheeky expression, surrounded by falling rose petals.

Portrait of a Native Hawaiian woman with dark skin and long black hair wearing a cream dress and pearl necklace, sitting at a dinner table with dessert and wine under a night sky.

A cosmic cat silhouette with deep black skin, glowing eyes, and nebula-like patterns glowing with stars set against a celestial cosmic background.

Prompts recommandés

score_9,score_8_up,score_7_up,score_6_up,score_5_up,score_4_up,source_anime,BREAK

score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, score_explicit, Art by wagashi \(dagashiya\), BREAK source_anime, bb \(swimsuit mooncancer\) \(second ascension\) \(fate\), jeanne d'arc alter \(fate\), purple hair, white hair, purple eyes, yellow eyes, long hair, 1girl, ribbon, ahoge, hair ribbon, full nelson, doggystyle, sex, uncensored, (changing room:1.2) background

Prompts négatifs recommandés

3d, monochrome, signature, torn clothes

Paramètres recommandés

samplers

Euler a

steps

cfg

resolution

1216x832, 832x1216

vae

sdxl_vae.safetensors (235745af8d)

Paramètres haute résolution recommandés

denoising strength

0.45

Conseils

For better compatibility with certain LoRAs train on AutismMix_pony version.

Add '3d' to negative prompts for a more traditional anime style.

Avoid using schizo negatives and SDXL/pony negative embeds as they may worsen results.

If issues arise running the model, use the webui: https://github.com/lllyasviel/stable-diffusion-webui-forge

Use the extension https://github.com/hako-mikan/sd-webui-prevent-artifact if you encounter noise outputs.

Sponsors du créateur

If you have any issues running this model, try using this web UI: https://github.com/lllyasviel/stable-diffusion-webui-forge

For noise reduction: https://github.com/hako-mikan/sd-webui-prevent-artifact

Mix of pony with some stuff. It's an attempt at making pony more predictable and less dependent on schizo negatives without removing its comprehension and artist knowledge. Personally I'm using AutismMix_confetti for general use and AutismMix_pony for certain loras. If you want to train a lora on top of autism I recommend doing so in the AutismMix_pony version for better compatibility. The Lightning versions require specific settings to work, read the "about model" information under download.

What is the difference between the models:

AutismMix_confetti is a small amount of AnimeConfettiTune and AutismMix_pony. Has less style swing than pony and better hands. I prefer this one.

AutismMix_pony is a merge of ponyv6 with loras, its more compatible with certain styles made for the base ponydiffusion model.

AutismMix_DPO is AutismMix_confetti+DPO lora, made by request. Very similar to confetti version.

Add 3d to negs if you want a more traditional anime style. Quality tags should be same as ponyv6, but feel free to experiment: "score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, source_anime, BREAK"

From my testing schizo negatives and those negative embeds made for SDXL/pony make it worse, but do whatever you want.

If you have any issues running this model I suggest using this webui: https://github.com/lllyasviel/stable-diffusion-webui-forge

As well as this extension if you get noise outputs:
https://github.com/hako-mikan/sd-webui-prevent-artifact